t: adapt character translation helpers to not use Perl

We have a couple of helper functions that translate characters, e.g.
from LF to NUL or NUL to 'Q' and vice versa. These helpers use Perl
scripts, but they can be trivially adapted to instead use tr(1).

Note that one specialty here is the handling of NUL characters in tr(1),
which historically wasn't implemented correctly on all platforms. But
quoting tr(1p):

    It was considered that automatically stripping NUL characters from
    the input was not correct functionality.  However, the removal of -n
    in a later proposal does not remove the requirement that tr
    correctly process NUL characters in its input stream.

So when tr(1) is implemented following the POSIX standard then it is
expected to handle the transliteration of NUL just fine.

Refactor the helpers accordingly, which allows a bunch of tests to pass
when Perl is not available.

Signed-off-by: Patrick Steinhardt <[email protected]>
Signed-off-by: Junio C Hamano <[email protected]>

Author: pks-t

t/test-lib-functions.sh

@@ -88,15 +88,15 @@ test_decode_color () {
 }
 
 lf_to_nul () {
-	perl -pe 'y/\012/\000/'
+	tr '\012' '\000'
 }
 
 nul_to_q () {
-	perl -pe 'y/\000/Q/'
+	tr '\000' 'Q'
 }
 
 q_to_nul () {
-	perl -pe 'y/Q/\000/'
+	tr 'Q' '\000'
 }
 
 q_to_cr () {